Is standardization of the PSA cutpoint for biochemical recurrence after surgery a good idea?


A paper just published in this month’s issue of the Journal of Urology has suggested that the true cutpoint for PSA levels indicative of biochemical recurrence in the highest number of patients is actually 0.4 ng/ml (as opposed to the oft-stated 0.2 ng/ml).

The problem, of course, is that there is no standard PSA level that can be used to accurately define biochemical recurrence. In fact (as we shall see below) there is actually very good reason for not trying to use a “one size fits all” PSA-based definition of biochemical recurrence!

The paper by Toussi et al. that proposes “consideration” of the use of a PSA level of 0.4 ng/ml or greater as the standard definition of biochemical recurrence after radical prostatectomy is based on their analysis of data from 13,512 patients with cT1-2N0M0 prostate cancer, all of whom underwent radical prostatectomy at the Mayo Clinic between 1987 and 2010.  They used these data to examine the utility of various single PSA cutpoints (0.2, 0.3, 0.4 and 0.5 ng/ml or greater) and also of two other confirmatory PSA value definitions (0.2 ng/ml or greater followed by PSA levels > 0.2 ng/ml and 0.4 ng/ml or greater followed by PSA levels of > 0.4 ng/ml).

What Toussi et al. found can be summarized as follows:

  • The average (median) postoperative follow-up on their patients was 9.1 years.
  • With that median follow-up period
    • A detectable post-surgical PSA developed in 5,041 patients (37.3 percent).
    • Systemic prostate cancer progression developed in 512 patients (3.8 percent).
  • After reaching a PSA cutpoint of 0.2 ng/ml, 61 percent of patients exhibited a continued PSA increase over the next 5 years.
  • After reaching a PSA cutpoint of 0.3 ng/ml, 67 percent of patients exhibited a continued PSA increase over the next 5 years.
  • After reaching a PSA cutpoint 0.4 ng/ml, 74 percent of patients exhibited a continued PSA increase over the next 5 years, plateauing at that level of 0.4 ng/ml.
  • The strongest association between biochemical recurrence and systemic progression occurred using a single PSA cutpoint of 0.4 ng/ml or greater.

However, in an accompanying editorial in the Journal of Urology, Tosoian and Pierorazio note carefully that not all biochemical recurrences are equal in type or likelihood. Specifically, they point out that a man who starts out as a high-risk patient and who is confirmed as such by the surgical and post-surgical pathology findings is much more likely to show early signs of recurrence that need early, second-line treatment than a man who starts out as a low-risk patient and whose low risk level is again confirmed by the surgical and post-surgical pathology findings. In the latter patients, it may be quite reasonable to delay any second-line therapy as long as one can in order to see if his PSA stabilizes again (this avoiding over-treatment of this patient). In the former patient, in contrast, one probably wants to be initiating second-line treatment as soon as possible in order to maximize the opportunity for a cure.

Tosoian and Pierorazio refer to an earlier paper by Mir et al. in the journal European Urology. In that paper, Mir et al. use careful statistical analysis to show exactly why it is entirely reasonable to use differing PSA cutpoints to validate the needs for second-line treatment of men with a rising PSA after initial surgery (depending on specific risk factors). Tosoian and Pierorazio go on to conclude that what we really need are studies to scrutinize with care the accuracy of differing PSA cutpoints for recurrence in differing patient groups carefully stratified by risk. It occurs to your sitemaster that there are probably several large patient series available (at places like Johns Hopkins, the Mayo Clinic, Memorial Sloan-Kettering Cancer Center, the University of California San Francisco, and elsewhere) that could be used to develop and validate such a set of differing, risk-stratified PSA cut points for recurrence based on patient risk category — at least until we can find a better test for evidence of clinically significant prostate cancer recurrence.

Standardization of clinical practice based on medical test data is a very good idea some of the time … but humans don’t tend to accommodate well to the idea that “one size fits all”. And using a PSA level of 0.4 ng/ml as a way to determine whether all men with prostate cancer with a rising PSA have actually got recurrent disease would seem to be a very good example of a not very good idea (especially from the perspective of high-risk patients who are hoping to avoid second-line therapy but also hoping to make quite sure their cancer doesn’t metastasize unnecessarily).

Editorial comment: The “New” Prostate Cancer InfoLink thanks Dr. Jeff Tosoian for promptly providing us with the full text of the editorial he and Dr. Pierorazio had published in the Journal of Urology.

6 Responses

  1. Their analysis seems to set an upper limit on how high the PSA should ever be allowed to get before considering salvage therapy. 3/4 patients who reached 0.4 ng/ml had detectable systemic progression within 5 years, indicating that 0.4 is a clear signal that the cancer had already progressed too far. But it seems to me that the whole point of setting a biochemical recurrence cutpoint is to find the lowest PSA that reliably signals clinical local recurrence. We want to start salvage treatment at the earliest time to avoid systemic progression, while not treating so early that many men would be over-treated.

    Other retrospective analyses imply that we can comfortably wait for a PSA of at least 0.03 before considering early salvage therapy. As the Sitemaster points out, this remains a challenging decision.

  2. I know this study is based on surgery to remove the prostate, but what, if any, takeaways would apply to BCR after HIFU treatment? I received HIFU in February and my initial post-HIFU PSA is 0.58, which both my Florida HIFU oncologist and my local oncologist (who recommended surgery) agreed was a good result. My prostate was not removed but ablated, i.e., “cooked” in place. The HIFU doc said that the <0.2 PSA can achieved by HIFU only through aggressive treatment of nerve bundles and urethra, which he spared.

  3. Dear David:

    It is impossible to use these data to make any decisions about PSA cutpoints for BCR after HIFU (or any other type of treatment). Surgery is the only form of treatment in which the goal is to remove the entire prostate and related tissues (the seminal vesicles) whose cells are the primary manufacturer of PSA. In every other form of treatment one is only ablating these tissues in some manner, and there are always going to be at least a few prostate cells (even if no prostate cancer cells) left behind afterwards.

  4. David,

    I think you are raising an important point about focal ablation. We do not yet know how to biochemically (using PSA) monitor progression afterwards. After surgery, we know that “undetectable” PSA connotes success. After primary radiation, we’ve learned by experience that a rise of 2 points from the lowest PSA alerts us that further investigation is warranted. Because the amount of tissue ablated with focal ablation varies from patient to patient, I doubt that we will ever have a single PSA increase that everyone agrees raises red flags. Therefore, men who are treated with focal ablation have to be on a kind of active surveillance. This means tracking PSA and having periodic biopsies to monitor progression.

  5. With the removal of all organs associated with producing PSA, there should be no PSA evident and the PSA level should be less than 0.1 ng/ml with usual testing, and, for those who have ultrasensitive testing, the PSA level should more likely be less than 0.01ng/ml. It is true that there are other sources that can produce near insignificant PSA; however, even then the PSA level after surgical removal should be no more than 0.03 ng/ml. I would personally see any PSA level of 0.1 ng/ml or higher with normal testing after surgical removal as reason for concern and close attention with further regular PSA testing.

  6. Dear Chuck:

    I don’t think anyone is suggesting that one shouldn’t monitor (with care) the PSA levels of anyone with a rising PSA post-surgery. The issue is only “When does one intervene?”

    We know that it is often impossible to remove “all” prostate tissue at the time of radical prostatectomy — usually because of anatomic consideration and the desire on the part of the patient and/or the surgeon to retain good quality urinary and erectile function./ This means that small amounts of prostate tissue (if not prostate cancer tissue) will remain in the body. No wise patient will want to have follow-up radiation therapy if there is no more cancer to be eliminated. But his PSA may be rising. This means that the question of when to intervene with salvage radiation therapy could be a big deal, especially in the cases of men who started out with low-volume, intermediate-risk prostate cancer that may well all have been removed at the time of surgery.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: